Paper of the Month: October 2022

Once a month during the academic year, the statistics faculty select a paper for our students to read and discuss. Papers are selected based on their impact or historical value, or because they contain useful techniques or results.

Ghosh, M., Sinha, B. K. and Mukhopadhyay, N. (1976). Multivariate sequential point estimation. Journal of Multivariate Analysis 6: 281-294.

Wednesday, Oct. 12, 2022
12:15-1:00 p.m.
Austin Building, Room 326

Notes Preparer: Nitis Mukhopadhyay, Professor of Statistics

Distinguished Chair Professor, Dr. Malay Ghosh, from the Department of Statistics, University of Florida-Gainesville, will be filmed on October 12, 2022 at UConn-Storrs for safe-keeping of the film at the American Statistical Association’s archive. I am thrilled by his selection. As his first Ph.D. advisee, I feel delighted that I will get yet another opportunity to watch-hear-cherish whatever Dr. Ghosh may decide to share from his wonderful life and influential career with the rest of us.

Dr. Ghosh has made fundamental contributions in practically all conceivable areas of statistical science. He is truly an all-rounder, always moving from one area to another seamlessly. Professor Dey suggested that I prepare an introductory seminar highlighting some landmark contributions of Dr. Ghosh in sequential analysis (SQA).

From many decades of collaborations with Dr. Ghosh, three publications in SQA come to mind. I have added full references below. The first one deals with multivariate analysis. The second one handles a nonparametric estimation problem. The third one introduces the notion of asymptotic second-order efficiency in sequential analysis. By glancing at these papers, one will gain a broad insight into Dr. Ghosh’s masterful approaches to research. During my in-person presentation, I will indicate some of the stories behind these collaborations and how they came to fruition. I hope to share glimpses of some elegant mixing of techniques from inference, decision theory, math stat, linear models, probabilistic tools, and large-sample methods. Papers [1]-[3] charted new discoveries with fresh and original directions in sequential estimation. I would hope to connect these three papers lightly and yet highlight some of the fundamental ideas from them which continue to draw researchers after these many years.

Additional Citations